Automatic tongue image segmentation based on gradient vector flow and region merging

Ning, Jifeng; Zhang, David; Wu, Chengke; Feng, Yuanyuan

doi:10.1007/s00521-010-0484-3

Cited by 66 publications

(79 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We further compared the proposed GaborFM method with three other tongue body segmentation methods: BEDC [35], Watershed [36], and PolarSnake [32]. Figure 10 shows the segmentation results of three typical tongue images obtained using the four segmentation methods: BEDC [35], Watershed [36], PolarSnake [32], and GaborFM.…”

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

“…Figure 10 shows the segmentation results of three typical tongue images obtained using the four segmentation methods: BEDC [35], Watershed [36], PolarSnake [32], and GaborFM. In BEDC [35], the bi-elliptical deformable template was adopted for contour initialization.…”

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

“…Since the shapes of the tongue body contour varies with tongue images, some contours could not be well approximated by bielliptical templates, and BEDC [35] obtained poor contour initialization results, as shown in Figure 10b,c. Watershed [36] adopted the region merging strategy for contour initialization, and this usually led to a nonsmooth contour, and over-and under-segmentation, as shown in Figure 10d,f. The PolarSnake [32] was designed for detecting dark lines, and location bias occurred for the detection of boundary pixels such as Points 3 and 4 in Figure 2.…”

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

“…Tongue body segmentation usually involves three steps: edge enhancement, contour initialization, and segmentation. For edge enhancement, conventional edge detectors, such as the Gaussian gradient [35], color gradient [36], and Canny detector [7], have been used. Considering the shape and gray-level characteristics of tongue contour, Zuo et al [10,11] proposed a polar edge detector to suppress adverse interference from the lip boundary and tongue fissures.…”

Section: Introductionmentioning

confidence: 99%

“…Considering the shape and gray-level characteristics of tongue contour, Zuo et al [10,11] proposed a polar edge detector to suppress adverse interference from the lip boundary and tongue fissures. For contour initialization, several heuristic [10], deformable model-based [35], and watershed-based [36] methods have been suggested. For segmentation, active contour models (ACM) [11,13,35,37] and the gradient vector flow (GVF) snake [38] have been used for the final segmentation of the tongue body [14,39,40].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Cui

Zhang

et al. 2013

EURASIP J. Adv. Signal Process.

Self Cite

View full text Add to dashboard Cite

Tongue body segmentation is a prerequisite to tongue image analysis and has recently received considerable attention. The existing tongue body segmentation methods usually involve two key steps: edge detection and active contour model (ACM)-based segmentation. However, conventional edge detectors cannot faithfully detect the contour of the tongue body, and the initialization of ACM suffers from the edge discontinuity problem. To address these issues, we proposed a novel tongue body segmentation method, GaborFM, which initializes ACM by performing fast marching over the two-dimensional (2D) Gabor magnitude domain of the tongue images. For the enhancement of the contour of the tongue body, we used the 2D Gabor magnitude-based detector. To cope with the edge discontinuity problem, the fast marching method was utilized to connect the discontinuous contour segments, resulting in a closed and continuous tongue body contour for subsequent ACM-based segmentation. Qualitative and quantitative results showed that GaborFM is superior to the other methods for tongue body segmentation.

show abstract

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

Section: Comparison With the Other Segmentation Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Cui

Zhang

et al. 2013

EURASIP J. Adv. Signal Process.

Self Cite

View full text Add to dashboard Cite

show abstract

Active contours driven by normalized local image fitting energy

Peng

Liu

2013

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARY In this paper, a new local region‐based active contour model that is driven by normalized local image fitting energy is proposed. By considering the local image fitting energy to extract the local image information, our model can effectively and efficiently segment images with intensity inhomogeneities. In addition, to keep the smoothness of the level set function, the time‐consuming reinitialization step widely adopted in traditional level set methods can be avoided by introducing a Gaussian filtering process. Experimental results on synthetic and real images demonstrate that the proposed model has much less CPU time and sensitivity to the initial contour than the well‐known local binary fitting model. Copyright © 2013 John Wiley & Sons, Ltd.

show abstract

A camera calibration method based on two orthogonal vanishing points

Zhao

Ning

2013

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARY A camera calibration method based on the two vanishing points, which are orthogonal to each other, is presented in this paper. On the basis of the facts that the modulus constraint can provide an equation and that two equations can be obtained from two pairs of vanishing points, which are orthogonal to each other, the self‐calibration is quasilinearly realized. The experiment results with simulate data and real image sequences show that the presented method is efficient. Copyright © 2013 John Wiley & Sons, Ltd.

show abstract

Automatic tongue image segmentation based on gradient vector flow and region merging

Cited by 66 publications

References 21 publications

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Fast marching over the 2D Gabor magnitude domain for tongue body segmentation

Active contours driven by normalized local image fitting energy

A camera calibration method based on two orthogonal vanishing points

Contact Info

Product

Resources

About